| // 3D Dom viewer, copy-paste this into your console to visualise the DOM as a stack of solid blocks. | |
| // You can also minify and save it as a bookmarklet (https://www.freecodecamp.org/news/what-are-bookmarklets/) | |
| (() => { | |
| const SHOW_SIDES = false; // color sides of DOM nodes? | |
| const COLOR_SURFACE = true; // color tops of DOM nodes? | |
| const COLOR_RANDOM = false; // randomise color? | |
| const COLOR_HUE = 190; // hue in HSL (https://hslpicker.com) | |
| const MAX_ROTATION = 180; // set to 360 to rotate all the way round | |
| const THICKNESS = 20; // thickness of layers | |
| const DISTANCE = 10000; // ¯\\_(ツ)_/¯ |
aka what i did to get from nothing to done.
note: these are designed to be primarily a re-install guide for myself (writing things down helps me memorize the knowledge), as such don't take any of this on blind faith - some areas are well tested and the docs are very robust, some items, less so). YMMV
Yoav Goldberg, April 2023.
With the release of the ChatGPT model and followup large language models (LLMs), there was a lot of discussion of the importance of "RLHF training", that is, "reinforcement learning from human feedback". I was puzzled for a while as to why RL (Reinforcement Learning) is better than learning from demonstrations (a.k.a supervised learning) for training language models. Shouldn't learning from demonstrations (or, in language model terminology "instruction fine tuning", learning to immitate human written answers) be sufficient? I came up with a theoretical argument that was somewhat convincing. But I came to realize there is an additional argumment which not only supports the case of RL training, but also requires it, in particular for models like ChatGPT. This additional argument is spelled out in (the first half of) a talk by John Schulman from OpenAI. This post pretty much
| <div style="width:100%; font-family:serif;font-size:16px;text-align:justify;"> | |
| <div style="font-weight:900;text-align:center;font-size:48px;border-bottom:3px solid black;font-variant:small-caps;margin-bottom:0px;"> | |
| The Eggbug Times | |
| </div> | |
| <div style="font-style:italic; font-size:18px;text-align:center;"> | |
| the fourth website to ever exist | |
| </div> | |
| <div style="text-transform:uppercase;font-size:48px;text-align:center;"> |
| #!/usr/bin/awk -f | |
| # This program is a copy of guff, a plot device. https://github.com/silentbicycle/guff | |
| # My copy here is written in awk instead of C, has no compelling benefit. | |
| # Public domain. @thingskatedid | |
| # Run as awk -v x=xyz ... or env variables for stuff? | |
| # Assumptions: the data is evenly spaced along the x-axis | |
| # TODO: moving average |
The following are appendices from Optics By Example, a comprehensive guide to optics from beginner to advanced! If you like the content below, there's plenty more where that came from; pick up the book!
| default['sshd']['sshd_config']['AuthenticationMethods'] = 'publickey,keyboard-interactive:pam' | |
| default['sshd']['sshd_config']['ChallengeResponseAuthentication'] = 'yes' | |
| default['sshd']['sshd_config']['PasswordAuthentication'] = 'no' |
All other guides I've seen (https://github.com/drduh/YubiKey-Guide being the most prolific) tell you to use the Yubikey's smartcard (PKCS#11) features with GnuPG via gpg-agent.
STOP THE MADNESS!
OpenSSH has supported OpenSC since version 5.4. This means that all you need to do is install the OpenSC library and tell SSH to use that library as your identity.
I largely followed Florin's blog post, but have a few notes to add regarding issues I encountered:
gpg-agent only automatically starts when gpg is used; for ssh, you'll need to ensure it's running.list command inside gpg --edit-card, look for the Key attributes line to see what is currently selected. On my YubiKey 4, it defaulted to 2048 bits for all keys:Key attributes ...: rsa2048 rsa2048 rsa2048
There isn't a way to configure DND for the weekend on Slack. However, you can use the /dnd 24 hours command to disable notifications for 24 hours using the [undocumented API][api] and [legacy tokens][token]. We're going to use the undocumented api and IFTTT to automate this.
DAYM93S0N from acme.slack.com/messages/DAYM93S0N/convo/GBG661YR2-1536698414.000100/)xoxp-4030334043-4959593942-3030320302032-sdf23rfasdf23rsf and CHANNEL looks like DAYM93S0N)
https://slack.com/api/chat.command?token=xoxp-4030334043-4959593942-3030320302032-sdf23rfasdf23rsf&channel=DAYM93S0N&command=%2Fdnd&text=24%20hours